On November 4, I had published on this blog a schematic outline of a "tree" at folio 204v of the selfsame manuscript. This is a plot, which I originally prepared for my own book but later decided not to use:
This figure, drawn in 1140 or thereabouts, is adapted from the century-old Stemma of Cunigunde, a drawing made in or shortly after 1013 when Cunigunde was anointed Holy Roman Empress. You can examine a reconstruction of that stemma on my website.
What does it show? The most important person here is R - Charlemagne - whose empire was divided among three of his grandsons by the Treaty of Verdun.
To enhance the chronicle, Ekkehard (or his predecessor editor Frutolf) repurposed the old diagram as a study in saintly ancestry by adding to it images of Arnulph and his holy mother Begga. Other manuscripts of the Ekkehard Chronicle present the Stemma of Cunigunde more or less faithfully, but the scribe-artist of the Erlangen codex decided to have some fun with it. He inverted it, and drew the figure of Arnulph at the left and Arnulph's saintly mother Begga at right. The bottom roundel (A in my plot) represents Arnulph.
Curiously, this artist omitted Cunigunde, although she had been the motive for creating the original drawing and she was revered in the entire Bamberg area, where this miniature was almost certainly made, as a holy figure and foundress of the cathedral. One must at least consider the possibility that the stemma was inverted in order to conceal her deliberate exclusion.
You can now enjoy the original at fol. 204v of the digital surrogate:
What change in medieval culture had made this startling inversion of the stemma not just possible, but acceptable to the customer, probably the Cistercian Monastery of Heilsbronn in Germany which became the long-term owner of this codex? Is this quirky conversion on an artist's desk the precise moment when the family tree, later to become a prestigious badge of nobility, was invented?
As with all big questions, the answer is not a simple one. A long inquiry was conducted into these issues by Christiane Klapisch-Zuber. As a historian of Renaissance culture, she was curious about the roots of the craze from the 15th to the 19th centuries to depict European aristocratic genealogies by painting vast leafy trees where portraits of ancestors were pinned to a trunk and out onto the boughs.
Researching her 2000 book, L'Ombre des Ancêtres, she cast far back into the medieval period, seeking precursors to those trees. This Heilsbronn tree of the late 12th century, as well as a couple of other painted trees made at roughly the same time at Weingarten, a monastery in the southwest of Germany, only dimly foreshadow the Renaissance craze.
The Weingarten artist, working between the years 1185 and 1191, drew a leafy inverted stemma of the powerful Welf family with its most ancient known ancestor peeking out from inside the trunk at ground level, while a wide space was reserved in the crown of the foliage to be occupied by the Welfs' most illustrious offspring (by female descent), King Frederick Barbarossa. This is in Fulda, 100 D.11, folio 13v and is online. The other from Weingarten is now lost but an image of the tree was published by its owner, the collector and dealer Robert Forrer, in 1907. Unfortunately that book is not yet online. In Europe, it does not enter the public domain until 2018.
The cultural change that took place in the 12th century with the rise of Gothic art was in fact much broader. Trees enjoyed a wide variety of uses in the graphic arts, ranging from trees of sevens as mnemonic devices to the tree manifestations of the Virga Jesse motif. Gothic cathedrals are in a certain sense trees of stone. This was what drove the experiments at Heilsbronn and Weingarten.
As I have already pointed out, inverted stemmata made to resemble trees with roots in soil are a rarity before the 16th century. It was 16th-century scholars like Scipione Ammirato who deserve the credit as the true originators of the family tree, not the medieval artists who created trees of ancestry more or less by fluke.